Saturday, December 14, 2013

Tycho Brahe: birthday

Another anniversary: Andrei Sakharov died on this day in 1989. A Russian dissident, a mastermind of their H-bomb, peace activist, and the author of the conditions needed for baryogenesis.
Today, the Google Doodle Booble reminds us of Tycho Brahe:

His name has appeared in nine TRF blog posts so this is the tenth one.

Tycho Brahe was born on December 14th, 1546, to a top-tier aristocratic family in Scania, Denmark (now South Sweden). He would attend a Latin school (age: 6-12 years; Latin is a subject for babies, of course), then study philosophy and rhetoric in Copenhagen (age: 13-16; typical subjects for children) and then the law school in Leipzig, DDR. When he was 19, he inherited lots of money so he had tons of time for his hobbies – alchemy and especially astronomy.

According to his memoirs, it's been decided for him to become a scholar when his uncle borrowed a 2-year-old Tycho and (without any parents' permission) remolded the baby into a scholar. As a rich guy, he would add the study of chemistry in Augsburg. In the 1580s, Tycho would possess about 1% of Denmark.

When he was 25, his dad died and Tycho returned to Denmark where he acquired his own observatory. A year later, he observed the the SN1572 supernova and described the observations in his "On the new star" essay. He would travel all over Europe for a while. But he was given a great offer by Frederick II of Denmark, the king, who built two observatories for Tycho.

This generosity may have been due to more than the king's respect to Tycho's talents; Tycho's uncle (and step father) has previously saved the king's life while drowning and sacrificed his own life (after pneumonia).

Tycho Brahe would work on his Danish observatories for 20 years but then he had a clash with the new king, Christian IV, so he left his homeland again. He would travel all over Europe for a while again. When he was 53, he got another major offer. He was invited by our (Austrian) emperor Rudolph II to Prague, the capital of the Austria Empire since 1583, and became the court astrologer. Unfortunately, less than two years of his life were ahead of him at that moment.

This offer was probably masterminded by Tadeáš Hájek of Hájek (Smallgrove of Smallgrove), a prominent Czech astronomer and mathematician and the personal physician of the emperor. I can't resist to emphasize that the end of the 16th century and the beginning of the 17th century was another proud (in my optics) period of my homeland in which our belonging to the Western European civilization was beyond any dispute. And I am confident that the Czech protestant aristocrats' loss in 1620 (the Battle of the White Mountain) was a dark event not only for the Czech patriots but for everyone who lived in Bohemia and Moravia (including the Germans); out lands were moved to the periphery of Western Europe, kind of.

Rudolph II (the guy on the "photograph" above where he appears as Vertumnus, the Roman God of seasons; (C) 1591–1592, Giuseppe Arcimboldo; Swedish queen Christina sent a batalion of thieves and terrorists to Prague in 1648 who successfully stole the painting along with 759 other paintings, 100 bronze statues, 33,000 coins and medallions, 600 pieces of crystal, 300 scientific instruments, manuscripts, and books, and she dared to call this robbery "peace": just if you wanted to know how Sweden may be capable of funding scientists who aren't very good – to compare, the British queen can't afford a few nuts for her bodyguards) really enjoyed arts, sciences, and protosciences. Prague would be full of these things. Due to his mysterious links to alchemy and astrology, there exist many legends – many of which are myths. For example, it's often said that the funny little Golden Lane near the Prague Castle with tiny houses is called in this way because the alchemists were working there to produce gold (folks even call it "Alchemists' Alley"). In reality, alchemists have never lived over there. Rudolph II is also said to be the ultimate controller of the Golem the Jewish soil robot. ;-)

Back to Brahe. He would build a new observatory in Benátky nad Jizerou (Venice Upon Iser: I really hope that at least someone enjoys all these semi-funny yet insightful translations of some of the Czech names), a town North of Prague. Brahe would be the boss of the observatory. During the last months of his life, John Kepler would be his assistant.

In 2010, a possible murder of Tycho Brahe was investigated in Prague where his remains were taken out of the resting place. Two years later, the conclusion was published: the amount of mercury in Brahe's blood wasn't enough to back the poisoning hypothesis (that's why a possibly career-driven Kepler the Killer has regained his apparent innocence, much like the obnoxious and suspicious king Christian IV) and no other credible method of murder had been proposed.

So the default belief is still that he has suffered from some kidney-related disease such as acute uremia. The problems could have been caused or helped by a massive dinner in the chateaux of Peter Vok of Rožmberk; Tycho Brahe had a painful bladder and couldn't pee right after this dinner, the sources suggest. However, he still went to a more ordinary dinner with the emperor. The legend says that it was impolite to use the restroom before the emperor (if you're an emperor, you just don't want some rank-and-file aristocrats to piss upon your toilet) which led to the near-explosion of his bladder.

Tycho Brahe would believe some kind of "enlightened geocentrism". He was stuck with geocentrism but all the other features of his model were modern and data-driven. He would prove that the celestial sphere wasn't constant. It was believed that all the temporary lights in the sky had to be comets or meteors – nearby objects. He would prove that a new star had no measurable parallax so it had to be very far, thus falsifying the particular formulation of the "constant sky" paradigm.

Most importantly, I think, Tycho's meticulous observations of the planetary motion were sort of needed for Kepler to induce his (Kepler's) laws – something that Isaac Newton was later able to derive from his new theory of mechanics and gravity. In some sense, Tycho represented the kind of a scientist that every other aristocrat could become with a sufficient amount of patience, hard work, and motivation. I wouldn't say that he's proven he was ingenious but his contributions to science were still extremely important.


  1. Thanks for your reply, Dilaton, and exactly, I like that you noticed the tension with the basics or "lore" that is already being explained in the context of T-duality.

    I received an e-mail reply from the authors but haven't read it yet....

  2. No way, Giotis, you suffer from a major misconception.

    The AdS/CFT is an exact duality that holds for any finite values of g and N.

    For large N or large g^2.N ('t Hooft coupling), some approximations of the bulk theory like SUGRA or the planar limit of the string theory may become good and corrections are relatively tiny. But if one is ready to include all the corrections from the finite 1/N effects etc., the two sides will exactly match for all finite values, too.

  3. Hi, I haven't banned him yet because he hasn't crossed the critical threshold of the "aggressiveness plus/times the amount of nonsense he is posting per unit time".

  4. Oh, I almost suspected that they are not string theorists, because I simply could not imagine string theorists being that confused about basics needed to understant the topics at hand ... ;-)

    Will you write an update after reading the mail? I at least would be interested in it ... :-)

  5. Dear Dilaton. I also received an e-mail from a very active researcher into the information loss issues who also thinks that the *ry's arguments don't make much sense.

    I also think it's OK (and perhaps even welcome) to reprint the e-mail from the *ry authors here:

    Dear Lubos,

    Thank you for your nice email, and for advertising our paper. Unfortunately, we don't agree with your interpretation of our paper. You seem to raise several different points, but probably what we disagree the most on is the existence of some form of "subregion duality".

    Before addressing that point directly, we don't see how our argument applies to any of the other examples you raised. On the boundary, you have an explicit factorization of the Hilbert space into two non-interacting subfactors. This is a very special statement. There must still exist some equivalent factorization of the bulk degrees of freedom; of course, it may not be obvious or local in the bulk, but we are unclear of what you think that decomposition looks like in the bulk even approximately.

    Our argument, which is similar in spirit to several previous authors' arguments (for instance, you might wish to see Marolf and Wall's paper), relies on a form of subregion duality (an idea which is also already present in the literature, see references in our paper). All we really need, though, is the statement that the CFT dual description of the insertion of a local operator in the bulk very close to one boundary only involves the (single) corresponding CFT. If you disagree with this statement, then we have a problem; but then, we also do not know what you mean when you say "Each exterior region is linked to a single dual CFT on one of the boundaries".

    Best Regards,
    Steve and Borun

  6. kashyap vasavadaDec 14, 2013, 1:22:00 PM

    Thanks. I see your point. As you can see I know very little about ST. So I am trying to get some birds' eye view of the status of ST. Some blogs are already rejecting it. I would like to get a fair balanced perspective without knowing technical details if that is possible, I do recognize that the opponents do not have any alternative as of today!

  7. I think there is a misunderstanding here…

    What I’m saying is that according to AdS/CFT the gravity dual of a SYM in the Coulomb branch is a D3 brane separated from the stack of the N D3 branes.

    But *only* in the large N approximation you can model this by putting a *probe*D3 in the gravitational background of the N D3s which in
    the near horizon limit is AdS5xS5 (this what Schwarz does i.e. he puts a probe
    on AdS5xS5); otherwise you have backreaction …

  8. Huh ok how could ArXiV ever accept that?

    Is this any better than stating that Type IIA String Theory can't be T-Dual to the Type IIB because the formfer is chiral, and the latter is non0c-hiral ?

  9. Huh, I don't understand why you don't want to go into the technical details.

    Why isn't it possible for you to get a standard book like those suggested by Lubos in his post on books in String Theory, like BBS for example?

  10. I think that there are much worse papers on the arXiv. ;-)

  11. kashyap vasavadaDec 14, 2013, 1:38:00 PM

    Good advice. But one excuse is that I am 76! The other excuse is that I did my Ph.D in theoretical physics in ancient times when most of this stuff was not around. But will try to understand this stuff gradually!

  12. Yes, you have a backreaction, but what's your problem with that (except for Sabine's Backreaction where the problems are understandable)?

    The exact effective action is supposed to incorporate all corrections from the backreaction, right?

  13. Excellent, I thought you were 17 or so.

  14. kashyap vasavadaDec 14, 2013, 1:46:00 PM

    Oh! I wish. I would give anything to be 17 again!!!

  15. Huh, how is that a valid excuse?

    See the 8th result here:

  16. kashyap vasavadaDec 14, 2013, 2:23:00 PM

    Hi! This link does not work.

  17. kashyap vasavadaDec 14, 2013, 2:32:00 PM

    Sorry, it did work finally.

  18. What exactly do you mean by the last sentence?
    As far as I have obseved, oponent blogs and the corresponding commenters there are simply not knowledgeable about string theory at a technical level. They have therefore neither the knowledge needed nor any legitimate right to judge this part of theoretical physics, even if they feel entitled to attack it in a very arrogant and pompous manner.

    The too widespreed opinion of aggressive dilettants and knownothings that they have the right to judge all kind of things they have no clue about, does not make their overreaching misbihavior legitimate, neither on blogs, popular media, or Physics SE ... !

  19. I think we are running in circles…

    Let me go to my original argument:

    AdS/CFT is saying that you can use the DBI action of a probe D3 brane in AdS5xS5 background to capture the low energy physics of the U(1)s in the broken phase of YM U(N) x U(1) *only* in the large N approximation.

    Now Schwarz is saying that you can do that for N=1 i.e. the Coulomb branch of U(2) and get the exact (or highly) effective action.

    So my original question was:

    How is that possible? AdS/CFT indicates that this (i.e. use the probe DBI on AdS etc) does not work for small N.

    You responded by saying that you don’t understand the paper and you see it as a straight forward implication of AdS/CFT.

    But as I have explained above this is not the case in my view. So I have to assert that this conjecture is something novel, not related AdS/CFT.

  20. Dear Giotis, together, we are running in circles because whenever I run in the right direction and stop running, you start to run in the opposite, wrong direction.

    There's *nothing* in AdS/CFT correspondence that would completely prevent one from accurately discussing physics of anything at a finite N - large N is *always* just about the possibility to ignore some corrections - and that's the last thing I will tel you again before I will encourage you to freely run in the straight, albeit completely wrong, direction. Run, baby, run. ;-)

  21. and by Susskind update the V1 of each that are discussed in an unhelpful fashion by these authors. In his latest he remains cautious saying that one cannot rule out firewalls. But in his discussions it seems that one can and must rule them out on arguments already advanced in the various of Lubos posts. I would call these generally arguments based on uncertainty that preclude as below and previously explained by Lubos and arguments based on QM complementarity holding because of er epr as a means of correlation but not allowing a signal to traverse as also well explained.
    I notice that in one case he would acknowledge it seems impossible to exclude firewalls but it seems to me that this is only so if we allow a signal to in fact cross the er-epr bridge and that would violate causality or we don't require uncertainty to include uncertainty of the interior mixing of the causal patches and that would violate Heisenberg. But he is far beyond my knowledge level in his discussions so
    I just don't know what the reservations he always expresses all mean to feel it reasonable to comment. I hope perhaps those of you who can will further develop this. I for one will be reading these papers and the earlier posts with great gusto for much of the rest of this season. Perhaps Lubos will give us a Christmas gift on this subject.

  22. Hello, Lumo,

    I have been greatly helped in coming to some sort of understanding about the ER=EPR concept through these pages. I do have one question that will probably show that sentence to be a fraud. When you measure one of a pair of entangled particles, what happens to the wormhole, and how? Does it pinch off, or is something odder going on?

  23. Dear Brian, thanks for your interest.

    If you measure the "complete information" (complete set of commuting observables) on one side of the wormhole, the outcome is a pure state for this side of the wormhole.

    This pure state is inevitably tensor-multiplied with a state on the other side - in other words, the entanglement goes to zero completely. The wormhole description is only smooth for highly (near-maximally) entangled states so for a low (or zero) entanglement, the wormhole is highly singular.

    You may say that the wormhole is still there just very thin or singular. You may also say that it's completely broken. It's the point of ER=EPR that there's no way to objectively distinguish between these two answers - the existence of a wormhole connection is not a good quantum observable (just like you can't define a linear operator that returns 1 for entangled states and 0 for non-entangled states). Both descriptions are possible - it's just about the choice of the basis.

    If you only measure a part of the information, you produce some errors that influence the life in the shared interior of the wormhole (black hole interiors of the two black holes that overlap). From the viewpoint of quantum information, the errors introduced by these "modest", partial measurements may be error-corrected.

    Be sure that there's no "bug" in ER=EPR that would be this simple. The hypothesis passes all such simple tests and much more.


  24. Thanks, Lumo! I wasn't looking for a bug - just insight.

  25. "All truth passes through 3 stages

    1st, it is ridiculed
    2nd, it is violently opposed.
    3rd, it is accepted as being self-evident"
    -- Arthur Schopenhauer, 19th century German philosopher